CME OnDemand: 2022 AOFAS Annual Meeting-Evaluation of Automated Coverage and Distance Mapping Selections to Improve Reliability and Clinical Utility of 3D Weightbearing CT Assessments

Evaluation of Automated Coverage and Distance Mapping Selections to Improve Reliability and Clinical Utility of 3D Weightbearing CT Assessments

Back to course

Pdf Summary

This study aimed to evaluate the use of automated selection processes to improve the clinical utility of 3D weightbearing CT (WBCT) assessments for progressive collapsing foot deformity (PCFD). PCFD is a complex 3D deformity that requires 3D analyses for accurate assessment. However, current analyses rely on manual identification of subregions, which limits their clinical usefulness. The objective was to assess an automated selection process in comparison to expert selections and enable clinical evaluation of subregional distance and coverage maps.<br /><br />The study analyzed WBCT data from 20 patients with PCFD and 10 control subjects retrospectively. Subregions of the peritalar surface were manually selected by two experts, and the same subregions were identified using an automated algorithm. A 3D distance mapping technique was used to create coverage maps across the peritalar surface. Comparisons between the automated selections and manual selections showed strong correlations, but average coverage was significantly higher for all subregions in the automated selections.<br /><br />The automated selections produced identical results with perfect intra-method correlations. The automated methods also replicated the gold standard manual selections with good reliability. This suggests that automated 3D assessments can aid in the diagnosis, staging, and objective evaluation of treatment effects for PCFD. The study concludes that these automated methods are well-suited for performing true 3D assessments of WBCT data in a clinical workflow, and that moving towards automated methods can improve reproducibility.<br /><br />However, there were no significant differences in mean distances in the middle or posterior facets between the automated and manual selections. Overall, the study demonstrates the reliability and replicability of automated methods for 3D WBCT assessments and suggests that automated methods should be adopted to improve reproducibility.

Asset Subtitle

Kevin N. Dibbern, PhD, Andrew Behrens, Nacime SB Mansur, MD, Kepler Carvalho, MD, Matthieu Lalevée, MD, Ki Chun Kim, MD, Eli Schmidt, Cesar De Cesar Netto, MD, PhD

Keywords

automated selection processes

clinical utility

3D weightbearing CT

progressive collapsing foot deformity

subregions

coverage maps

WBCT data

automated algorithm

reproducibility

objective evaluation